Till innehåll på sidan

Modeling Biochemical Network Involved In Striatal Dopamine Signaling

Tid: Fr 2018-03-02 kl 14.00

Plats: F3, Lindstedtsvägen 26, KTH Campus

Ämnesområde: Computer Science, Computational Biology

Respondent: Anu G. Nair , CST

Opponent: Asst. Professor Jakob Kisbye Dreyer

Handledare: Professor Jeanette Hellgren Kotaleski


In this thesis, I studied the molecular integration of reward-learning related neuromodulatory inputs by striatal medium-sized projection neurons (MSNs) using mass-action kinetic modeling.

It is known that, in reward learning, an unexpected reward results in transient elevation in dopamine (peak) whereas omission of an expected reward leads to transient dopamine decrease (dip). In silico experiments performed in the current study indicated that reward-related transient dopamine signals could act differentially on the cAMP/PKA signaling of the two MSN classes, D1 receptor expressing MSNs (D1 MSNs) and D2 receptor expressing MSNs (D2 MSNs). PKA in D1 MSN responded to dopamine peaks, whereas in D2 MSN it was affected by dopamine dips. Simulations further highlighted the possibility that cAMP/PKA signaling in D1 MSNs is tonically inhibited by acetylcholine by activating muscarinic M4 receptors under the basal condition. In this scenario, the D1 receptor activation by a dopamine peak does not have any downstream effect, unless the dopamine peak is accompanied by an acetylcholine dip that could release the M4-mediated inhibition. Such acetylcholine dips accompany dopamine peaks due to the time-locked dopaminergic bursts and cholinergic pauses observed in reward-learning. Thus, an acetylcholine dip could be viewed as a time window for dopamine signaling in D1 MSN. Similarly, the cAMP/PKA signaling in D2 MSN could be tonically inhibited by the dopamine-dependent D2 receptors. In this case, a dopamine dip results in the cAMP/PKA activation, and the strength of the downstream response depends on the level of basal adenosine, acting via A2a receptors. These results highlight how multiple neuromodulators could be integrated by striatal MSNs to produce effective downstream response. Such signal integration scenarios require that the dopamine and acetylcholine-triggered cAMP signaling be sufficiently powerful and sensitive. However, quantitative information regarding the efficacy of dopamine and acetylcholine on cAMP signaling is virtually nonexistent for living MSNs. Therefore, the effects of dopamine and acetylcholine on cAMP signaling were quantitatively characterized in this study by imaging genetically-encoded FRET-based biosensor expressed in mice brain slices. The measurements confirmed that the cAMP signaling in MSNs is quite sensitive and could strongly be influenced by neuromodulators, thus supporting the underlying model requirements, and thereby predictions.

Another parameter that is important for effective molecular signal integration is the relative timing between various convergent inputs. For example, studies have shown that LTP in D1 MSNs is produced if corticostriatal glutamate synaptic activity is shortly followed by a dopamine peak. However, there is no LTP if the order of the inputs is reversed. This temporal dependence is believed to result in various aspects of reward learning, such as reward causality, and is theoretically represented by the so-called eligibility trace. However, little is known how such temporal constraints emerge at the level of molecular signaling. I investigated the possible molecular mechanism responsible for the emergence of this temporal constraints, using computational modeling. This study proposes a novel molecular mechanism based on the coordinated activity of two striatally enriched phosphoproteins, DARPP-32 and ARPP-21 that could explain the emergence of the timing-dependence for postsynaptic signal integration, and thus a plausible molecular underpinning for the eligibility trace of reward learning.

In summary, the results presented in this thesis advance our understanding on how the striatal cAMP respond towards reward-related nueromodulator signals, and the downstream effects on synaptic signaling and reward learning.

The thesis in Diva